Conventional approaches for a distributed low probability of detection communications system with a large number of unique transmitters and a single or a few receivers, require receiver complexity proportional to the number of transmitters. To improve efficiency in terms of receiver complexity, two alternative designs are analyzed and compared to a reference receiver, whose complexity grows linearly as the number of transmitters increases. The first alternative system groups the transmitters into clusters whose pseudorandom noise codes have some chips in common. The resulting receiver would then perform two stages of processes: identification of the transmitting cluster and received bit detection. The total number of processes required for any given transmitter would be substantially less than the traditional receiver. The second alternative design would utilize a common long spreading code and a shorter cyclically shifted spreading code in each transmitter. The receiver utilizes the cyclic shift property of the fast Fourier transform to recover efficiently both the identity of active receivers and the data sent using a single branch. The complexity of the two proposed systems is compared to that of the reference system.

Approved for public release; distribution is unlimited.

http://archive.org/details/designofscalable109454577

US Navy (USN) author.